Java WeightedFragInfo类代码示例

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
/**
 * Fixes problems with broken analysis chains if positions and offsets are messed up that can lead to
 * {@link StringIndexOutOfBoundsException} in the {@link FastVectorHighlighter}
 */
public static WeightedFragInfo fixWeightedFragInfo(FieldMapper mapper, Field[] values, WeightedFragInfo fragInfo) {
    assert fragInfo != null : "FragInfo must not be null";
    assert mapper.fieldType().name().equals(values[0].name()) : "Expected FieldMapper for field " + values[0].name();
    if (!fragInfo.getSubInfos().isEmpty() && (containsBrokenAnalysis(mapper.fieldType().indexAnalyzer()))) {
        /* This is a special case where broken analysis like WDF is used for term-vector creation at index-time
         * which can potentially mess up the offsets. To prevent a SAIIOBException we need to resort
         * the fragments based on their offsets rather than using soley the positions as it is done in
         * the FastVectorHighlighter. Yet, this is really a lucene problem and should be fixed in lucene rather
         * than in this hack... aka. "we are are working on in!" */
        final List<SubInfo> subInfos = fragInfo.getSubInfos();
        CollectionUtil.introSort(subInfos, new Comparator<SubInfo>() {
            @Override
            public int compare(SubInfo o1, SubInfo o2) {
                int startOffset = o1.getTermsOffsets().get(0).getStartOffset();
                int startOffset2 = o2.getTermsOffsets().get(0).getStartOffset();
                return FragmentBuilderHelper.compare(startOffset, startOffset2);
            }
        });
        return new WeightedFragInfo(Math.min(fragInfo.getSubInfos().get(0).getTermsOffsets().get(0).getStartOffset(),
                fragInfo.getStartOffset()), fragInfo.getEndOffset(), subInfos, fragInfo.getTotalBoost());
    } else {
        return fragInfo;
    }
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
/**
 * Fixes problems with broken analysis chains if positions and offsets are messed up that can lead to
 * {@link StringIndexOutOfBoundsException} in the {@link FastVectorHighlighter}
 */
public static WeightedFragInfo fixWeightedFragInfo(FieldMapper mapper, Field[] values, WeightedFragInfo fragInfo) {
    assert fragInfo != null : "FragInfo must not be null";
    assert mapper.fieldType().names().indexName().equals(values[0].name()) : "Expected FieldMapper for field " + values[0].name();
    if (!fragInfo.getSubInfos().isEmpty() && (containsBrokenAnalysis(mapper.fieldType().indexAnalyzer()))) {
        /* This is a special case where broken analysis like WDF is used for term-vector creation at index-time
         * which can potentially mess up the offsets. To prevent a SAIIOBException we need to resort
         * the fragments based on their offsets rather than using soley the positions as it is done in
         * the FastVectorHighlighter. Yet, this is really a lucene problem and should be fixed in lucene rather
         * than in this hack... aka. "we are are working on in!" */
        final List<SubInfo> subInfos = fragInfo.getSubInfos();
        CollectionUtil.introSort(subInfos, new Comparator<SubInfo>() {
            @Override
            public int compare(SubInfo o1, SubInfo o2) {
                int startOffset = o1.getTermsOffsets().get(0).getStartOffset();
                int startOffset2 = o2.getTermsOffsets().get(0).getStartOffset();
                return FragmentBuilderHelper.compare(startOffset, startOffset2);
            }
        });
        return new WeightedFragInfo(Math.min(fragInfo.getSubInfos().get(0).getTermsOffsets().get(0).getStartOffset(),
                fragInfo.getStartOffset()), fragInfo.getEndOffset(), subInfos, fragInfo.getTotalBoost());
    } else {
        return fragInfo;
    }
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
protected String makeFragment( StringBuilder buffer, int[] index, Field[] values, WeightedFragInfo fragInfo,
    String[] preTags, String[] postTags, Encoder encoder ){
  StringBuilder fragment = new StringBuilder();
  final int s = fragInfo.getStartOffset();
  int[] modifiedStartOffset = { s };
  String src = getFragmentSourceMSO( buffer, index, values, s, fragInfo.getEndOffset(), modifiedStartOffset );
  int srcIndex = 0;
  for( SubInfo subInfo : fragInfo.getSubInfos() ){
    for( Toffs to : subInfo.getTermsOffsets() ){
      fragment
        .append( encoder.encodeText( src.substring( srcIndex, to.getStartOffset() - modifiedStartOffset[0] ) ) )
        .append( getPreTag( preTags, subInfo.getSeqnum() ) )
        .append( encoder.encodeText( src.substring( to.getStartOffset() - modifiedStartOffset[0], to.getEndOffset() - modifiedStartOffset[0] ) ) )
        .append( getPostTag( postTags, subInfo.getSeqnum() ) );
      srcIndex = to.getEndOffset() - modifiedStartOffset[0];
    }
  }
  fragment.append( encoder.encodeText( src.substring( srcIndex ) ) );
  return fragment.toString();
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
private void testCase( Query query, int fragCharSize, String expectedFragInfo,
    double expectedTotalSubInfoBoost ) throws Exception {
  makeIndexLongMV();

  FieldQuery fq = new FieldQuery( query, true, true );
  FieldTermStack stack = new FieldTermStack( reader, 0, F, fq );
  FieldPhraseList fpl = new FieldPhraseList( stack, fq );
  WeightedFragListBuilder wflb = new WeightedFragListBuilder();
  FieldFragList ffl = wflb.createFieldFragList( fpl, fragCharSize );
  assertEquals( 1, ffl.getFragInfos().size() );
  assertEquals( expectedFragInfo, ffl.getFragInfos().get( 0 ).toString() );

  float totalSubInfoBoost = 0;
  for ( WeightedFragInfo info : ffl.getFragInfos() ) {
    for ( SubInfo subInfo : info.getSubInfos() ) {
      totalSubInfoBoost += subInfo.getBoost();
    }
  }
  assertEquals( expectedTotalSubInfoBoost, totalSubInfoBoost, .0000001 );
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
public String[] createFragments( IndexReader reader, int docId,
    String fieldName, FieldFragList fieldFragList, int maxNumFragments,
    String[] preTags, String[] postTags, Encoder encoder ) throws IOException {

  if( maxNumFragments < 0 ) {
    throw new IllegalArgumentException( "maxNumFragments(" + maxNumFragments + ") must be positive number." );
  }

  List<WeightedFragInfo> fragInfos = fieldFragList.getFragInfos();
  Field[] values = getFields( reader, docId, fieldName );
  if( values.length == 0 ) {
    return null;
  }

  if (discreteMultiValueHighlighting && values.length > 1) {
    fragInfos = discreteMultiValueHighlighting(fragInfos, values);
  }

  fragInfos = getWeightedFragInfoList(fragInfos);
  int limitFragments = maxNumFragments < fragInfos.size() ? maxNumFragments : fragInfos.size();
  List<String> fragments = new ArrayList<>( limitFragments );

  StringBuilder buffer = new StringBuilder();
  int[] nextValueIndex = { 0 };
  for( int n = 0; n < limitFragments; n++ ){
    WeightedFragInfo fragInfo = fragInfos.get( n );
    fragments.add( makeFragment( buffer, nextValueIndex, values, fragInfo, preTags, postTags, encoder ) );
  }
  return fragments.toArray( new String[fragments.size()] );
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
public int compare( WeightedFragInfo o1, WeightedFragInfo o2 ) {
  if( o1.getTotalBoost() > o2.getTotalBoost() ) return -1;
  else if( o1.getTotalBoost() < o2.getTotalBoost() ) return 1;
  // if same score then check startOffset
  else{
    if( o1.getStartOffset() < o2.getStartOffset() ) return -1;
    else if( o1.getStartOffset() > o2.getStartOffset() ) return 1;
  }
  return 0;
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
public String[] createFragments( IndexReader reader, int docId,
    String fieldName, FieldFragList fieldFragList, int maxNumFragments,
    String[] preTags, String[] postTags, Encoder encoder ) throws IOException {

  if( maxNumFragments < 0 ) {
    throw new IllegalArgumentException( "maxNumFragments(" + maxNumFragments + ") must be positive number." );
  }

  List<WeightedFragInfo> fragInfos = fieldFragList.getFragInfos();
  Field[] values = getFields( reader, docId, fieldName );
  if( values.length == 0 ) {
    return null;
  }

  if (discreteMultiValueHighlighting && values.length > 1) {
    fragInfos = discreteMultiValueHighlighting(fragInfos, values);
  }

  fragInfos = getWeightedFragInfoList(fragInfos);
  int limitFragments = maxNumFragments < fragInfos.size() ? maxNumFragments : fragInfos.size();
  List<String> fragments = new ArrayList<String>( limitFragments );

  StringBuilder buffer = new StringBuilder();
  int[] nextValueIndex = { 0 };
  for( int n = 0; n < limitFragments; n++ ){
    WeightedFragInfo fragInfo = fragInfos.get( n );
    fragments.add( makeFragment( buffer, nextValueIndex, values, fragInfo, preTags, postTags, encoder ) );
  }
  return fragments.toArray( new String[fragments.size()] );
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
protected String makeFragment( StringBuilder buffer, int[] index, Field[] values, WeightedFragInfo fragInfo,
        String[] preTags, String[] postTags, Encoder encoder ){
    return super.makeFragment(buffer, index, values, FragmentBuilderHelper.fixWeightedFragInfo(mapper, values, fragInfo),
            preTags, postTags, encoder);
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
 protected String makeFragment( StringBuilder buffer, int[] index, Field[] values, WeightedFragInfo fragInfo,
         String[] preTags, String[] postTags, Encoder encoder ){
     return super.makeFragment(buffer, index, values, FragmentBuilderHelper.fixWeightedFragInfo(mapper, values, fragInfo),
             preTags, postTags, encoder);
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
protected String makeFragment( StringBuilder buffer, int[] index, Field[] values, WeightedFragInfo fragInfo,
        String[] preTags, String[] postTags, Encoder encoder ){
    return super.makeFragment(buffer, index, values, FragmentBuilderHelper.fixWeightedFragInfo(mapper, values, fragInfo), preTags, postTags, encoder);
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
@Override
 protected String makeFragment( StringBuilder buffer, int[] index, Field[] values, WeightedFragInfo fragInfo,
         String[] preTags, String[] postTags, Encoder encoder ){
     return super.makeFragment(buffer, index, values, FragmentBuilderHelper.fixWeightedFragInfo(mapper, values, fragInfo), preTags, postTags, encoder);
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
/**
 * do nothing. return the source list.
 */
@Override
public List<WeightedFragInfo> getWeightedFragInfoList( List<WeightedFragInfo> src ) {
  return src;
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
/**
 * Sort by score the list of WeightedFragInfo
 */
@Override
public List<WeightedFragInfo> getWeightedFragInfoList( List<WeightedFragInfo> src ) {
  Collections.sort( src, new ScoreComparator() );
  return src;
}
 

import org.apache.lucene.search.vectorhighlight.FieldFragList.WeightedFragInfo; //导入依赖的package包/类
public abstract List<WeightedFragInfo> getWeightedFragInfoList( List<WeightedFragInfo> src );